Recent research shows that social structures or non-sensitive attributes of users can increase risks of user sensitive attribute disclosure in social networks. Most of the existing private attribute anonymization schemes have many defects, such as lack of proper model, too much distortion on attributes distribution, neglect social structure and non-sensitive attributes' influence on sensitive attributes. In this paper, an attribute privacy preservation scheme based on node anatomy is proposed. It allocates original node's attribute links and social links to new nodes to improve original node's anonymity, thus protects user from sensitive attribute disclosure. Meanwhile, it measures social structure influence on attribute distribution, and splits attributes according to attributes' correlations. Experimental results show that the proposed scheme can maintain high data utility and resist private attribute disclosure.